|Publication number||US7774313 B1|
|Application number||US 11/289,148|
|Publication date||Aug 10, 2010|
|Priority date||Nov 29, 2005|
|Publication number||11289148, 289148, US 7774313 B1, US 7774313B1, US-B1-7774313, US7774313 B1, US7774313B1|
|Original Assignee||Symantec Corporation|
|Export Citation||BiBTeX, EndNote, RefMan|
|Patent Citations (12), Non-Patent Citations (4), Referenced by (3), Classifications (8), Legal Events (5)|
|External Links: USPTO, USPTO Assignment, Espacenet|
Businesses, residences, and other enterprises have come to rely on computing systems to access, generate and manage their key files, documents, and other operational data. Often, the data itself is many times more valuable to an enterprise than the computing hardware that stores the data. Accordingly, in this information age, many enterprises have taken precautions to protect their data.
One way of protecting data is to introduce storage redundancy. For example, a primary computing system maintains and operates upon the active data. Meanwhile, a backup computing system maintains a copy of the data. This may be accomplished by periodically taking a snapshot of the active data as that active data existed at the time of the snapshot, and by providing the snapshot copy of the protected data to a backup location on a periodic basis. For instances, snapshot-based backups often are configured to occur on a daily basis, though more or less frequent backup periods are also common. Should the active data be lost, the most recent backup copy of the data may be restored, thereby recovering the active data as the active data existed at the time of the last backup.
In some instances, however, it may be highly undesirable to lose all of the changes to the data that occurred since the time of the last snapshot backup. In such cases, Continuous Data Protection CDP may be more desirable. In CDP-based backup systems, an initial copy of the live data to be protected is captured and transferred to a backup medium. All subsequent writes to the live data are then time-stamped and journaled to the backup medium after each write operation. For instance, whenever a write operation occurs, the time, content, and target location of the write operation are documented in a data structure, which is then provided to the backup medium.
Since CDP-based backup systems back up and time stamp each block written to the protected data, it is possible to recover any state of the protected data since the time of the initial backup, in contrast to traditional snapshot-based backup systems which can recover only those states of the protected data that happen to exist at the time of the prior snapshot-based backup. This allows for a recovery back to any point in time to recover any version of a file.
Conventionally, CDP-based backup systems journal every block of every file modified on a volume, consuming large amounts of storage and requiring significant network bandwidth to replicate the journaled data to remote backup systems.
The foregoing problems with the prior state of the art are overcome by the principles of the present invention, which are directed towards policy-based performance of continuous data protection on protected data. A write request targeted to a portion of the protected data is detected. In addition, a journal policy is accessed. The journal policy represents policy for how frequently to journal write requests to a backup medium and/or what backup medium to journal write requests to depending on one or more characteristics of write request targets. The journaling policy data structure is then used to determine whether the write request should be presently journaled and/or to identify the backup medium that the write request should be journaled to based on the characteristics of the portion of the protected data targeted by the write request. The journaling policy may, but need not, be configured so as to preserve storage and/or network backwidth associated with the journaling process.
Additional embodiments the invention will be set forth in the description that follows, and in part will be obvious from the description, or may be learned by the practice of the invention. The embodiments of the invention may be realized and obtained by means of the instruments and combinations particularly pointed out in the appended claims. These and other embodiments of the present invention will become more fully apparent from the following description and appended claims, or may be learned by the practice of the invention as set forth hereinafter.
In order to describe the manner in which the above-recited and other advantages and features of the invention can be obtained, a more particular description of the invention briefly described above will be rendered by reference to specific embodiments thereof which are illustrated in the appended drawings. Understanding that these drawings depict only typical embodiments of the invention and are not therefore to be considered to be limiting of its scope, the invention will be described and explained with additional specificity and detail through the use of the accompanying drawings in which:
The principles of the present invention relate to policy-based performance of continuous data protection on protected data. First, a general computing system will be described with respect to
Computing systems are now increasingly taking a wide variety of forms. Computing systems may, for example, be handheld devices, appliances, laptop computers, desktop computers, mainframes, or distributed computing systems. In this description and in the claims, the term “computing system” is defined broadly as including any device or system (or combination thereof) that includes at least one processor, and a memory capable of having thereon computer-executable instructions that may be executed by the processor. The memory may take any form and may depend on the nature and form of the computing system. A computing system may be distributed over a network environment and may include multiple constituent computing systems.
As used herein, the term “module” or “component” can refer to software objects or routines that execute on the computing system. The different components, modules, engines, and services described herein may be implemented as objects or processes that execute on the computing system (e.g., as separate threads). While the system and methods described herein may be implemented in software, implementations in hardware, and in combinations of software and hardware are also possible and contemplated.
In the description that follows, embodiments of the invention are described with reference to acts that are performed by one or more computing systems. If such acts are implemented in software, one or more processors of the associated computing system that performs the act direct the operation of the computing system in response to having executed compute-executable instructions. An example of such an operation involves the manipulation of data. The computer-executable instructions (and the manipulated data) may be stored in the memory 104 of the computing system 100.
The phrase “computer-readable storage medium,” as referred to herein, generally refers to any form of storage device or medium capable of storing computer-readable instructions. Examples of computer-readable storage media include—without limitation—physical media, such as magnetic-storage media (e.g., hard disk drives and floppy disks), optical-storage media (e.g., CD- or DVD-ROMs), electronic-storage media (e.g., solid-state drives and flash media), and the like.
The phrase “computer-readable storage medium,” as referred to herein, generally refers to any form of physical storage device or medium capable of storing computer-readable instructions. Examples of computer-readable storage media include magnetic-storage media (e.g., hard disk drives and floppy disks), optical-storage media (e.g., CD- or DVD-ROMs), electronic-storage media (e.g., solid-state drives and flash media), and the like.
As previously mentioned, however, journaling of each write request can take significant storage and/or bandwidth. The method 200, however, reduces the storage and/or bandwidth requirements associated with performing continuous data protection by not necessarily journaling every write request, and/or by perhaps clustering groups of two or more write requests together before each journaling operation to the backup medium. The journaling decision may be made for each write request in response to a journal policy.
The method 200 may be performed by the computing system 100 of
At any time before, during and/or after the write request is received (act 201), a journaling policy data structure is accessed (act 202). In this description and in the claims, a computing system “accessing” a data structure means that all or perhaps even just a portion of the data structure is read to the extent that the computing system may then act upon the portion of the data structure read. The journaling policy data structure represents journaling policy for how frequently to journal write requests to a backup medium depending on one or more characteristics of write request targets. For example, if the target of the write request is a particular file, or portion of a file, the journal policy may specify a frequency for journaling based on a file type of the file.
For instance, an enterprise could configure the journaling policy such that file types that are more critical to the operation of the enterprise might be journaled more frequently or aggressively (e.g., 100% of the time) than are file types that are less critical to an organization. For instance, documents with a “.doc” extension type indicating a MICROSOFT Word document may be journaled whenever a write request is received, while MP3 audio files are journaled less frequently (perhaps 33% of the time). The journaling policy could also specify that journaling is to occur depending on certain content of the file, directory, or object being written to. For example, if the file being written to contains the words “urgent” or “confidential”, journaling may be done for each write request to the file, whereas if the file does not contain those words, perhaps journaling is only done half the time. Thus, content parameters of the target may be target characteristics that dictate how frequently to journal, whether the target be a directory, a file, an object, or a sub-file block. Other examples of content parameters include the number of times that a certain word appears in a target file. For instance, if the target document contains more than a certain number of instances of the word “discovery”, or if perhaps the word discovery appears more times than there are pages in the document, journaling might occur more frequently since the file might more likely be the object of discovery. A content parameter might also include the author or creator of a file. Those who are more likely to need more fine-grained recovery may have their associated files journaled more frequently.
Once the journaling policy is accessed (act 202), it is then determined whether the write request should be presently journaled using the journaling policy and based on one or more characteristics of the target portion of the protected data targeted by the write request (decision block 203). There is an infinite variety of different algorithms that could be employed using the principles of the present invention to make a decision on whether or not to presently journal a write request. One example is now given for illustrative purposes only.
Suppose, for example, a first write request is received, and the journaling policy assigns a journal frequency of 33% of the first write request since its target is an MP3 file. This amount is added to a sum corresponding to that file that is initially zero, leaving the file-specific sum at 33% after the addition. If the sum after the addition is less than 100%, then the write request is not journaled if the write request is at the file-level, or if the write request is at the sub-file block level (e.g., at the sector or cluster level) the write request is stored in a cache to thereby postpone the journaling of that write request. If the sum after the addition reaches or exceeds 100%, the current write request is journaled.
Network bandwidth and storage is preserved whether the write request is at the file system level or at the sub-file block level. For instance, if the write request tracking is performed at the file level, then not every file write request will always be cached. Only the most current file write request will be journaled when there is an automated decision to journal based on the journaling policy. Furthermore if the write request tracking is performed at the sub-file block level (e.g., at the sector or cluster level), the write request may be journaled in a single journaling data structure along with all other previously-postponed and cached write requests to sub-file blocks corresponding to the same file. However, if this collection of sub-file write request includes writes to the same sub-file block, only the latest write request to that sub-file block is journaled, thereby saving network bandwidth and backup storage.
After journaling for a particular file occurs, the sum for that file is then reinitialized to zero, or perhaps just decremented by 100%. Continuing with the example, suppose a second write request is received in which the journal policy assigns a frequency of 75%. The sum is then incremented by 75% to reach 108%. Now the sum exceeds 100% so the second write request is journaled. If the journaling is performed with other prior write requests, the collection is journaled in a single journal message, and with a single time stamp. The sum is then reinitialized to zero, or perhaps decremented by 100% to 8%, and the process continues for each subsequent write request.
Regardless of how the decision is made, if it is determined that the write request should be journaled (Yes in decision block 203), it is then determined whether or not there are one or more prior write requests for that file that are yet to be journaled to the backup medium (decision block 204). If the method 200 is performed at the file level, decision block 204 may be skipped proceeding directly to the journaling process (act 206) since only the most recent file write request will be journaled. If there are other write requests to be journaled (Yes in decision block 204), the current write request is journaled along with the one or more prior write requests whose journaling was postponed due to the journal policy (act 205). For instance, if the write requests are at the sub-file block level (e.g., at the cluster level), then if one of the cluster write requests is to be journaled, all other non-redundant cluster write requests will be journaled within the same journal message. For instance, if there are three write requests to the same cluster, only the last write request to that cluster is journaled. If there are no such prior write requests (No in decision block 204), the current write request is journaled alone in the journal message provided to the backup medium (act 206).
On the other hand, if the policy indicates that the write request is not to be journaled (No in decision block 203), then the write request is cached (act 207) for future journaling until one or more further write requests are detected if the write request is at the sub-file block level. Otherwise, if the write request were at the file level, the write request may even be discarded.
There are some write requests that may not be immediately journaled when the write request is first executed on the primary storage. In other cases, the file write request may never be journaled at all. Accordingly, it is possible that in the event the data in the primary storage is to be corrupted while there are still write requests pending for future journaling, that the recovery may not be completely up-to-date to the actual time of the corruption. However, the principles of the present invention allow policy setters (such as, for example, an Information Technology team at a corporation) to identify a more desirable balance between granularity of recovery versus expeditious use of storage and network resource given decision parameters such as write target file metadata and content when performing continuous data protection.
Continuing with the example of
Accordingly, the principles of the present invention permit journaling frequency to be determined based on a journaling policy. The journaling policy cold be used to make other journaling decisions as well. For example,
Once again, the method 400 as illustrated may be initiated upon detecting a write request targeted to a target portion of the protected data (act 401). Also, the journaling policy is accessed (act 402). In this case, however, the decision parameters of the journaling policy may correlate the identity of the target backup medium based on the file type, other file metadata, and/or file content of the file being written to.
The backup medium to be journaled to is identified (act 403) using the journaling policy and comparing the characteristics of the target file to the decision parameters in the journal policy. The file write request is then journaled to the identified backup medium (act 404).
As an illustrative example,
As previously described,
Accordingly, the principles of the present invention allow for policy-based journaling in automatically deciding how frequently and where to journal write requests when performing continuous data protection. This allows for the intelligent setting of journaling policy so as to balance backup granularity with efficient usage of storage and network bandwidth when performing continuous data protection.
The present invention may be embodied in other specific forms without departing from its spirit or essential characteristics. The described embodiments are to be considered in all respects only as illustrative and not restrictive. The scope of the invention is, therefore, indicated by the appended claims rather than by the foregoing description. All changes, which come within the meaning and range of equivalency of the claims, are to be embraced within their scope.
|Cited Patent||Filing date||Publication date||Applicant||Title|
|US5047918||Dec 19, 1988||Sep 10, 1991||Tektronix, Inc.||File management system|
|US5572709 *||Jun 18, 1993||Nov 5, 1996||Lucent Technologies Inc.||Using dynamically-linked libraries to add side effects to operations|
|US6941322 *||Apr 25, 2002||Sep 6, 2005||International Business Machines Corporation||Method for efficient recording and management of data changes to an object|
|US7225208||Sep 30, 2003||May 29, 2007||Iron Mountain Incorporated||Systems and methods for backing up data files|
|US7260564||Apr 6, 2001||Aug 21, 2007||Virage, Inc.||Network video guide and spidering|
|US20020174139 *||May 21, 2002||Nov 21, 2002||Christopher Midgley||Systems and methods for backing up data files|
|US20030061456 *||Jul 17, 2002||Mar 27, 2003||Yuval Ofek||Apparatus and methods for copying, backing up and restoring logical objects in a computer storage system by transferring blocks out of order or in parallel|
|US20030229650 *||Jun 6, 2002||Dec 11, 2003||International Business Machines Corporation||Method and apparatus for selective caching of transactions in a computer system|
|US20050193031||Mar 18, 2005||Sep 1, 2005||Livevault Corporation||Systems and methods for backing up data files|
|US20050210041 *||Mar 18, 2004||Sep 22, 2005||Hitachi, Ltd.||Management method for data retention|
|US20050235016 *||Apr 14, 2004||Oct 20, 2005||Takashi Amano||Method and apparatus for avoiding journal overflow on backup and recovery system using storage based journaling|
|US20070022117 *||Jul 21, 2005||Jan 25, 2007||Keohane Susann M||Accessing file system snapshots directly within a file system directory|
|1||Final Office Action received in U.S. Appl. No. 11/288,587; Apr. 13, 2009.|
|2||Final Office Action received in U.S. Appl. No. 11/288,587; Aug. 21, 2008.|
|3||Non-Final Office Action received in U.S. Appl. No. 11/288,587; Apr. 17, 2008.|
|4||Non-Final Office Action received in U.S. Appl. No. 11/288,587; Nov. 24, 2008.|
|Citing Patent||Filing date||Publication date||Applicant||Title|
|US8762341||Dec 22, 2011||Jun 24, 2014||Symantec Corporation||Efficiently configuring multiple backup data policies with information specifying data to backup|
|CN102193845A *||May 30, 2011||Sep 21, 2011||华中科技大学||Data recovery method|
|CN102193845B||May 30, 2011||Dec 19, 2012||华中科技大学||Data recovery method|
|U.S. Classification||707/640, 707/694|
|Cooperative Classification||G06F11/1471, G06F11/1435, G06F2201/84, G06F11/1464|
|Feb 3, 2006||AS||Assignment|
Owner name: SYMANTEC CORPORATION, CALIFORNIA
Free format text: ASSIGNMENT OF ASSIGNORS INTEREST;ASSIGNOR:NACHENBERG, CAREY STOVER;REEL/FRAME:017545/0264
Effective date: 20060207
|Jan 23, 2014||FPAY||Fee payment|
Year of fee payment: 4
|Feb 4, 2016||AS||Assignment|
Owner name: VERITAS US IP HOLDINGS LLC, CALIFORNIA
Free format text: ASSIGNMENT OF ASSIGNORS INTEREST;ASSIGNOR:SYMANTEC CORPORATION;REEL/FRAME:037697/0412
Effective date: 20160129
|Feb 23, 2016||AS||Assignment|
Owner name: BANK OF AMERICA, N.A., AS COLLATERAL AGENT, NORTH
Free format text: SECURITY INTEREST;ASSIGNOR:VERITAS US IP HOLDINGS LLC;REEL/FRAME:037891/0001
Effective date: 20160129
Owner name: WILMINGTON TRUST, NATIONAL ASSOCIATION, AS COLLATE
Free format text: SECURITY INTEREST;ASSIGNOR:VERITAS US IP HOLDINGS LLC;REEL/FRAME:037891/0726
Effective date: 20160129
|Apr 18, 2016||AS||Assignment|
Owner name: VERITAS TECHNOLOGIES LLC, CALIFORNIA
Free format text: MERGER AND CHANGE OF NAME;ASSIGNORS:VERITAS US IP HOLDINGS LLC;VERITAS TECHNOLOGIES LLC;REEL/FRAME:038455/0752
Effective date: 20160329